The Effectiveness of Nonverbal Symbolic Signs and Metaphors in Advertisements

Principle 1:
Symbolic signs require motivation and a non-trivial amount of cognitive load to process.
Symbolic signs are signs that have little or no resemblance to their physical meaning; they are metaphor, and their meaning is highly dependant on their context. An example would be an image of a few flowers: depending on the context, it could be interpreted as meaning 'spring', or 'feminine', or 'a field'. Unlike iconic signs, which physically resemble their meaning (and which are automatically recognised), symbolic signs require an amount of cognitive processing to decipher. The amount of cognitive processing performed depends on the motivation to decipher the sign. Therefore, unlike iconic signs, symbolic signs are unlikely to be deciphered by all who encounter them.

Principle 2:
Only consumers with 'moderate' motivation to decipher a sign will be persuaded by it.
Advertisers use symbolic signs to attribute non-utilitarian meaning to a product in order to change the perception of that product in consumers' minds. Consumers with low motivation to interpret the sign will not engage in the 'process of abduction', whereupon an inference is made as a result of seeing the sign. On the other hand, those with high motivation will also engage in a 'counterargument' after first deciphering the sign. An example of a counterargument is 'I know that the flower on the plane means that it will be sunny at my destination, but I also know it is a ruse to get me to use that airline.' Because of the tendancy to counterargue, it is difficult to persuade someone with high motivation using a symbolic sign. As it is known that deciphering a symbolic sign takes less germane load than counterarguing, only those with a moderate level of interest are likely to be persuaded by a symbolic sign's inference, as they will not make the extra effort required to counterargue.

Reference:
DeRosia, E. D. (2008). The Effectiveness of nonverbal symbolic signs and metaphors in advertisements: An experimental enquiry. Psychology & Marketing, 25(3): 298-316.